Chromatography Forum

We have been analyzing olive oil polar fractions (extract obtained from olive oil by 80% aqueous methanol) filtered through 0.22 um membrane and the C18 column (1.6 um, 2.0 mm, 75mm ShimPak XR-ODS III) progressively deteriorated. This was obvious by the shape of some peaks of interest. In addition some extra peaks appeared, late eluting and in retention times that interfere with the analysis (so they cannot be just ignored). When this happended another issue that happended was a leakage due to a rotten rubber part in the autosampler where the syringe is inserted.
Since then we obtained a new column but the peaks continued to elute with the gradient without making any injection even of a blank. We tried the washing of the system lines and the autosampler with solvents of diffferent polarity. Some peaks were removed but one still remains. We changed the cell, we excluded the autosampler, even the mixer after cleaning the solvent reservoirs, changing solvents and checking their UV-Vis spectra before their use but still this peak remains.
The mobile phase is a) water HPLC purity (0.2% phosphoric acid) additionally filtered through 0.45um membrane b) methanol HPLC gradient grade c) acetonitrile gradient grade. Through all the gradient the methanol and acetonitrile are always 1:1 v/v.
The interferences absorb only at 280 nm but are not fluorescent under the conditions of analysis (exc.280nm/em.320 nm).
Any idea that could help?

What types of maintenance and service have been done? Has the system been run through a PV or OQ to check operation? What you are describing sounds like a worn injector rotary valve seal (a normal maintenance item). It could also be due to not washing down the column after each run with a solvent system that is stronger than the one used in the method too (and which dissolves the oil).

Because of their hydrophobic nature, Olive oils are usually run with non-conventional detectors such as ELSD or CAD. Those allow the use of solvents which can fully dissolve the samples and get rid of the problem of contamination over time. My boss said he used to run them in mobile phases with methylene chloride all of the time using non-UV detectors for that reason.

You need stronger solvents more than just acidified Water and Methanol in your mobile phase! You can try either Methylene Chloride (normal phase) or IPA (reverse phase).

Thank you for the reply. We used also isopropanol but still the peak remains

Are you using only HPLC solvents? Are you ultrafiltering (minimum 0.22 um) the sample beforehand?

Fluorescence is prone to external contamination. Be sure ALL glassware is washed and rinsed with Acetone. Be sure ALL analysts are wearing gloves, since skin oils can fluoresce.

Good Morining,

If 100% IPA fails to work, perhaps mixing in 80:10:10 IPA/Water/Formic Acid or 80:10:10 IPA/Water/Ammonium Hydroxide for compounds more soluble in acids and bases, respectively, or 30:70 Phosphoric Acid/Water will work a bit better. Please don't connect either detector or a column to the LC while using any of these combinations, and run the effluent to waste directly...and flush these out of the system thoroughly with water after you're done cleaning.

Thank you for the suggestion. We tried today but it did not work. The peak remains and its size remain constant without seeing any decrease after washing the system.

Hi again,

To be clear on this end, cleaning with a 30:70 (v/v) mixture of H3PO4/HPLC water was executed without success? Was this mixture applied to the injector as well?

In any case, if the HPLC system can handle 5N HNO3, you could try this next--this was once commonly used to passivate the stainless steel capillaries used in HPLC equipment. I'd ask the vendor of the HPLC first before trying this out.

Given the lack of success of both organic solvents and inorganic acids up till now, perhaps a rebuild of the system is in order--as well as some changes made to the routine maintenance of the HPLC equipment in between uses as suggested above.